Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.arch > #25041 > unrolled thread

article wrt black holes and fractals out of chaotic systems...

Back to article view | Back to comp.arch

Contents

  article wrt black holes and fractals out of chaotic systems... "Chris M. Thomasson" <no@spam.invalid> - 2014-11-08 15:12 -0800
    Re: article wrt black holes and fractals out of chaotic systems... "Chris M. Thomasson" <no@spam.invalid> - 2014-11-09 22:03 -0800
    Re: article wrt black holes and fractals out of chaotic systems... "Chris M. Thomasson" <no@spam.invalid> - 2014-12-10 20:55 -0800
      Re: article wrt black holes and fractals out of chaotic systems... "Chris M. Thomasson" <no@spam.invalid> - 2014-12-10 21:09 -0800
        Re: article wrt black holes and fractals out of chaotic systems... Terje Mathisen <terje.mathisen@tmsw.no> - 2014-12-11 18:39 +0100
        Re: article wrt black holes and fractals out of chaotic systems... Raymond Wiker <rwiker@gmail.com> - 2014-12-12 20:13 +0100
          Re: article wrt black holes and fractals out of chaotic systems... Michael Barry <barrym95838@yahoo.com> - 2014-12-14 12:49 -0800
        Re: article wrt black holes and fractals out of chaotic systems... "Christoph M. Becker" <cmbecker69@arcor.de> - 2014-12-12 20:35 +0100
        Re: article wrt black holes and fractals out of chaotic systems... "Chris M. Thomasson" <no@spam.invalid> - 2014-12-12 17:52 -0800
          Re: article wrt black holes and fractals out of chaotic systems... "James Van Buskirk" <not_valid@comcast.net> - 2014-12-13 11:52 -0700
            Re: article wrt black holes and fractals out of chaotic systems... "Chris M. Thomasson" <no@spam.invalid> - 2014-12-15 13:58 -0800
            Re: article wrt black holes and fractals out of chaotic systems... "Chris M. Thomasson" <no@spam.invalid> - 2014-12-15 14:02 -0800
              Re: article wrt black holes and fractals out of chaotic systems... "James Van Buskirk" <not_valid@comcast.net> - 2014-12-15 16:58 -0700
        Re: article wrt black holes and fractals out of chaotic systems... Melzzzzz <mel@zzzzz.com> - 2014-12-13 04:39 +0100
      Re: article wrt black holes and fractals out of chaotic systems... edprochak@gmail.com - 2014-12-12 11:05 -0800
        Re: article wrt black holes and fractals out of chaotic systems... "Chris M. Thomasson" <no@spam.invalid> - 2014-12-12 18:00 -0800

#25041 — article wrt black holes and fractals out of chaotic systems...

FWIW, an article wrt black holes and fractals out of chaotic systems:

http://www.nature.com/news/black-hole-mergers-cast-kaleidoscope-of-shadows-1.16283

I am trying to detect fractals wrt observing/intercepting
computer simulated gravitational waves, but so far, I
can only get an idea of the overall border... Not the
internals of the fractal. Here is an example vector field:

https://plus.google.com/101799841244447089430/posts/cdJWGvgPMnd

The border seems to come through, but the "internal"
spirals seem damn near impossible to detect from
observations far outside the field over vast regions of
space, along with other unrelated clutter wrt the target. 

[toc] | [next] | [standalone]

#25073

> "Chris M. Thomasson"  wrote in message 
> news:m3m81j$8gh$1@speranza.aioe.org...

> FWIW, an article wrt black holes and fractals out of chaotic systems:

[...]

I was wondering if this to too terribly
off topic as to cancel immediately, or,
if I post in a "sane" manner, and not
create new topics and stick to my own
tree...

?



<off-topic>
Here is an animation:

https://www.youtube.com/watch?v=hOUUCk3Gdg4

I did wrt a vector field. 10 static points,
and one animated along a parametric spiral
starting at the center of the 10 static
points.

Now since the field can be computed at any
point, why couldn't I use the GPU via pixel
shaders to gain gradients across every pixel.
Banging head on wall, and thinking of every
point in the plane as definite "candidate" for
a pixel in the vector field as a whole.
</off-topic>


Does this sound Kosher in any way, shape or
form?

;^o 

[toc] | [prev] | [next] | [standalone]

#25645

FWIW, here is a crude animated GPU-based tour of Julia set around
the main cardioid of the Mandelbrot set. This is implemented in
a pixel shader that has real time color cycling, animation and
point-and-click zooming capabilities. I am planning to extend
this out into a full blown online fractal explorer, complete with
a custom function parser. I am wondering if it works for you?

So far, I can get it to work in Chrome and Firefox. IE does not
seem to want to render; I am probably doing something slightly
wrong in the shader, but it does work in these other browsers.
Perhaps IE does not like the fact that I am breaking out of a for
loop. IIRC, the shader language does not like breaking out of a
loop early; Humm...

Anyway, here is a link to the online program:

http://webpages.charter.net/appcore/fractal/webglx/ct_complex_field.html

This is using an escape method I came up with that actually
generates an entire infinite vector field, complete with field lines.
Last time I saw an algorithm for field lines in a complex escape time
fractal, the method was just way too complex; mine is almost too easy!

FWIW, it is fairly similar to some of my simulated 2d gravitational field
line animation's on YouTube:

https://www.youtube.com/watch?v=hOUUCk3Gdg4

https://www.youtube.com/watch?v=h0WEsw1F0xY


I think I found a unique method to generate a vector field that totally
comprises everything out of the main set; so to speak.

Well, does it work for you?



Thanks!



BTW, I am using 32-bit floats, so the distance you can zoom to before rapid
pixilation will be rather limited!

Sorry about that for I have not emulated 64-bit floats with 32-bit base
floats yet, but it is on the list!

:^)

[toc] | [prev] | [next] | [standalone]

#25646

(cross-posted to comp.lang.javascript; I forgot to add this
newsgroup before I posted it here!   :^(   )

"Chris M. Thomasson"  wrote in message 
news:m6b84v$o6j$1@speranza.aioe.org...

FWIW, here is a crude animated GPU-based tour of Julia set around
the main cardioid of the Mandelbrot set. This is implemented in
a pixel shader that has real time color cycling, animation and
point-and-click zooming capabilities. I am planning to extend
this out into a full blown online fractal explorer, complete with
a custom function parser. I am wondering if it works for you?

So far, I can get it to work in Chrome and Firefox. IE does not
seem to want to render; I am probably doing something slightly
wrong in the shader, but it does work in these other browsers.
Perhaps IE does not like the fact that I am breaking out of a for
loop. IIRC, the shader language does not like breaking out of a
loop early; Humm...

Anyway, here is a link to the online program:

http://webpages.charter.net/appcore/fractal/webglx/ct_complex_field.html

This is using an escape method I came up with that actually
generates an entire infinite vector field, complete with field lines.
Last time I saw an algorithm for field lines in a complex escape time
fractal, the method was just way too complex; mine is almost too easy!

FWIW, it is fairly similar to some of my simulated 2d gravitational field
line animation's on YouTube:

https://www.youtube.com/watch?v=hOUUCk3Gdg4

https://www.youtube.com/watch?v=h0WEsw1F0xY


I think I found a unique method to generate a vector field that totally
comprises everything out of the main set; so to speak.

Well, does it work for you?



Thanks!



BTW, I am using 32-bit floats, so the distance you can zoom to before rapid
pixilation will be rather limited!

Sorry about that for I have not emulated 64-bit floats with 32-bit base
floats yet, but it is on the list!

:^) 

[toc] | [prev] | [next] | [standalone]

#25659

Chris M. Thomasson wrote:
> (cross-posted to comp.lang.javascript; I forgot to add this
> newsgroup before I posted it here!   :^(   )
>
> "Chris M. Thomasson"  wrote in message
> news:m6b84v$o6j$1@speranza.aioe.org...
>
> FWIW, here is a crude animated GPU-based tour of Julia set around
> the main cardioid of the Mandelbrot set. This is implemented in
> a pixel shader that has real time color cycling, animation and
> point-and-click zooming capabilities. I am planning to extend
> this out into a full blown online fractal explorer, complete with
> a custom function parser. I am wondering if it works for you?

I'm using SeaMonkey (similar to Firefox), it works very nicely indeed!

Color me impressed. :-)

Terje
>
> So far, I can get it to work in Chrome and Firefox. IE does not
> seem to want to render; I am probably doing something slightly
> wrong in the shader, but it does work in these other browsers.
> Perhaps IE does not like the fact that I am breaking out of a for
> loop. IIRC, the shader language does not like breaking out of a
> loop early; Humm...
>
> Anyway, here is a link to the online program:
>
> http://webpages.charter.net/appcore/fractal/webglx/ct_complex_field.html
>
> This is using an escape method I came up with that actually
> generates an entire infinite vector field, complete with field lines.
> Last time I saw an algorithm for field lines in a complex escape time
> fractal, the method was just way too complex; mine is almost too easy!
>
> FWIW, it is fairly similar to some of my simulated 2d gravitational field
> line animation's on YouTube:
>
> https://www.youtube.com/watch?v=hOUUCk3Gdg4
>
> https://www.youtube.com/watch?v=h0WEsw1F0xY
>
>
> I think I found a unique method to generate a vector field that totally
> comprises everything out of the main set; so to speak.
>
> Well, does it work for you?
>
>
>
> Thanks!
>
>
>
> BTW, I am using 32-bit floats, so the distance you can zoom to before rapid
> pixilation will be rather limited!
>
> Sorry about that for I have not emulated 64-bit floats with 32-bit base
> floats yet, but it is on the list!
>
> :^)


-- 
- <Terje.Mathisen at tmsw.no>
"almost all programming can be viewed as an exercise in caching"

[toc] | [prev] | [next] | [standalone]

#25687

"Chris M. Thomasson" <no@spam.invalid> writes:

> ... 
>
> http://webpages.charter.net/appcore/fractal/webglx/ct_complex_field.html
>

Beautiful.

[toc] | [prev] | [next] | [standalone]

#25744

On Friday, December 12, 2014 11:14:00 AM UTC-8, Raymond Wiker wrote:
> "Chris M. Thomasson" <no@spam.invalid> writes:
> 
> > ... 
> >
> > http://webpages.charter.net/appcore/fractal/webglx/ct_complex_field.html
> >
> 
> Beautiful.

Very nice.  I'm getting almost three frames/sec on my old bone-stock Dell Inspiron 545 running full-screen Chrome on Vista.

Mike

[toc] | [prev] | [next] | [standalone]

#25693

Chris M. Thomasson wrote:

> http://webpages.charter.net/appcore/fractal/webglx/ct_complex_field.html
> 
> Well, does it work for you?

My Chrome 39.0.2171.71 m (64-bit) alerts: "Cannot load WegGL
Context!!!!".  The same happens after having updated to 39.0.2171.95 m
(64-bit).

[F'up2 comp.lang.javascript]

-- 
Christoph M. Becker

[toc] | [prev] | [next] | [standalone]

#25710

> "Chris M. Thomasson"  wrote in message 
> news:m6b8u1$pjk$1@speranza.aioe.org...

> FWIW, here is a crude animated GPU-based tour of Julia set around
> the main cardioid of the Mandelbrot set.

[...]

Thank you all for the very kind comments! I still have a
lot of work to do on this. For instance, I need to add a
"throttle" because some people have told me that it runs
a bit too fast on their high-end graphics cards.

:^) 

[toc] | [prev] | [next] | [standalone]

#25722

"Chris M. Thomasson"  wrote in message 
news:m6g64k$8o9$1@speranza.aioe.org...

> "Chris M. Thomasson"  wrote in message
> > news:m6b8u1$pjk$1@speranza.aioe.org...

> > FWIW, here is a crude animated GPU-based tour of Julia set around
> > the main cardioid of the Mandelbrot set.

> [...]

> Thank you all for the very kind comments! I still have a
> lot of work to do on this. For instance, I need to add a
> "throttle" because some people have told me that it runs
> a bit too fast on their high-end graphics cards.

Also it has to detect window size and fill the graphics
window it's being drawn in, maybe disable scaling if in
full screen mode. Works MSIE 11.

[toc] | [prev] | [next] | [standalone]

#25766

> "James Van Buskirk"  wrote in message news:m6i1to$eq5$1@dont-email.me... 

> > "Chris M. Thomasson"  wrote in message 
> > news:m6g64k$8o9$1@speranza.aioe.org...

> > "Chris M. Thomasson"  wrote in message
> > > news:m6b8u1$pjk$1@speranza.aioe.org...

> > > FWIW, here is a crude animated GPU-based tour of Julia set around
> > > the main cardioid of the Mandelbrot set.

> > [...]

> > Thank you all for the very kind comments! I still have a
> > lot of work to do on this. For instance, I need to add a
> > "throttle" because some people have told me that it runs
> > a bit too fast on their high-end graphics cards.

> Also it has to detect window size and fill the graphics
> window it's being drawn in, maybe disable scaling if in
> full screen mode. Works MSIE 11.

I agree, and thank you for giving it a try James! However,
I must say that I am a bit worried that a User just might not
have a GPU that is "up to par" wrt the size of their "viewport"
and end up destroying any peformance they may have had at
the smaller size...

Humm... You can set this manually right now by saving
the HTML locally, then edit it using your favorite text
editor. Just open the HTML, and search for the text
"prv_resize_canvas" without the quotes, and manually
change the dimensions; they are hardcoded at 1024x1024
now. I can make it an "automatic" process, but am a little
afraid!

;^o


FWIW, I recently witnessed this running on a friends computer
equipped with a very nice NIVIDA graphics card, and it almost
runs a little too fast for my eye! So, this would work nicely there,
but perhaps not everywhere.

Humm...


I have a lot to ponder here!

:^)

[toc] | [prev] | [next] | [standalone]

#25768

> "James Van Buskirk"  wrote in message news:m6i1to$eq5$1@dont-email.me... 

> > "Chris M. Thomasson"  wrote in message 
> > news:m6g64k$8o9$1@speranza.aioe.org...
> > [...]

> Also it has to detect window size and fill the graphics
> window it's being drawn in, maybe disable scaling if in
> full screen mode. Works MSIE 11.

FWIW, it does not want to work on my MSIE 11! It just sits
there with a blank canvas. As to why it does not work for me,
well, I still do not know!

;^(


Thank you for trying it anyway James.

[toc] | [prev] | [next] | [standalone]

#25770

"Chris M. Thomasson"  wrote in message 
news:m6nlp6$fei$1@speranza.aioe.org...

> "James Van Buskirk"  wrote in message news:m6i1to$eq5$1@dont-email.me...

> > "Chris M. Thomasson"  wrote in message 
> > news:m6g64k$8o9$1@speranza.aioe.org...
> > [...]

> > Also it has to detect window size and fill the graphics
> > window it's being drawn in, maybe disable scaling if in
> > full screen mode. Works MSIE 11.

> FWIW, it does not want to work on my MSIE 11! It just sits
> there with a blank canvas. As to why it does not work for me,
> well, I still do not know!

The strange thing is, when I open it from the link in your original post,
MSIE 11 opens it without problems, but when I open it by clicking on
it from File Explorer, I get a popup that says Internet Explorer
restricted this webpage from running scripts or Active X controls.
When I press the Allow blocked content button, it then runs.
Changes:

function prv_resize_canvas(canvas_)
{
    //canvas_.width = 1024;
    canvas_.width = 3840;
    //canvas_.height = 1024;
    canvas_.height = canvas_.width * (9.0 / 16.0);
}

I turned MSIE zoom down to 100% and disabled Windows 8.1 scaling,
and now it fills the screen after I hit F11.  At this resolution the 
animation is more
of an ooze than the annoying flashing that it was at 1024 X 1024.  Still 
some
terrain is wasted because the caption covers the whole top of the image
rather that a neat little box and the scroll bars, but it's getting there.

[toc] | [prev] | [next] | [standalone]

#25714

On Wed, 10 Dec 2014 21:09:03 -0800
"Chris M. Thomasson" <no@spam.invalid> wrote:

> 
> http://webpages.charter.net/appcore/fractal/webglx/ct_complex_field.html
> 
> 
> Well, does it work for you?
> 
> 
> 
> Thanks!

Works, I see animation in firefox.

> 
> 
> 

[toc] | [prev] | [next] | [standalone]

#25685

On Wednesday, December 10, 2014 11:56:03 PM UTC-5, Chris M. Thomasson wrote:
> FWIW, here is a crude animated GPU-based tour of Julia set around
> the main cardioid of the Mandelbrot set. This is implemented in
> a pixel shader that has real time color cycling, animation and
> point-and-click zooming capabilities. I am planning to extend
> this out into a full blown online fractal explorer, complete with
> a custom function parser. I am wondering if it works for you?
> 
> So far, I can get it to work in Chrome and Firefox. IE does not
> seem to want to render; I am probably doing something slightly
> wrong in the shader, but it does work in these other browsers.
> Perhaps IE does not like the fact that I am breaking out of a for
> loop. IIRC, the shader language does not like breaking out of a
> loop early; Humm...
> 
> Anyway, here is a link to the online program:
> 
> http://webpages.charter.net/appcore/fractal/webglx/ct_complex_field.html
> 
[]
> 
> Well, does it work for you?
> 
> 
Hypnotic. Very cool.
It does remind me of doing fractals on a 1Mhz Comodore64 and waiting 24hours for the image to render. 8^)
> 
> Thanks!
> 
No, Thank you!!


Is the image generation running on your machine and just streaming the image to the browser? Or is it running in the browser. (I didn't check the page source. I am just so fascinated by the rapid image updates.

Thanks again.

[toc] | [prev] | [next] | [standalone]

#25711

> wrote in message 
> news:19885f32-9e5e-431c-b713-6bbff88e0f82@googlegroups.com...

> > On Wednesday, December 10, 2014 11:56:03 PM UTC-5, Chris M. Thomasson 
> > wrote:
> > FWIW, here is a crude animated GPU-based tour of Julia set around
> > the main cardioid of the Mandelbrot set.
[...]
> > Anyway, here is a link to the online program:
> >
> > http://webpages.charter.net/appcore/fractal/webglx/ct_complex_field.html
> >
> > Well, does it work for you?

> Hypnotic. Very cool.
> It does remind me of doing fractals on a 1Mhz Comodore64
> and waiting 24hours for the image to render. 8^)

I remember doing fractals on my Apple IIGS; waiting was basically
a given.

;^)



> > Thanks!

> No, Thank you!!

You are very welcome! I am pleased that it works for you.

:^D



> Is the image generation running on your machine and just streaming
> the image to the browser? Or is it running in the browser. (I didn't
> check the page source. I am just so fascinated by the rapid image
> updates.

The rendering is running in the GPU while the Julia point generation is
running in the CPU. It is a completely local process. AFAICT, this would
simply not be possible without using the graphics hardware!

:^o 

[toc] | [prev] | [standalone]

Back to top | Article view | comp.arch

csiph-web